Fungal Diversity

Fungal latent pathogens and endophytes from leaves of ()

Angel Romero\ Gloria Carrion*l and Victor Rico-Gray2

IDepartamento Hongos, 2Departamento de Ecologia Vegetal, lnstituto de Ecologia, A.C. Apdo. 63, Xalapa, VER 91000, Mexico; *e-mail: [email protected]

Romero, A., Carrion, G. and Rico-Gray, V. (2001). Fungal latent pathogens and endophytes from leaves of Parthenium hysterophorus (Asteraceae). Fungal Diversity 7: 81-87.

Endophytic fungi were isolated from leaves of the weed Parthenium hysterophorus in order to establish whether the endophytes were the same fungi as had previously been recorded on scenescent or diseased leaf tissues. Seven surface sterilization methods were used. Alternaria zinniae, A. helianthi, Cylindrocarpon sp., Curvularia brachyspora, Fusarium sp., Nigrospora oryzae, Penicillium funiculosum and Periconia sp. were isolated The methods used to isolate endophytes may represent a tool for the identification of biological control agents of weeds.

Key words: biological control, endophytes, weeds.

Introduction Parthenium hysterophorus (Asteraceae: ) is a native herbaceous from tropical and subtropical America (Nash and William, 1976). In Australia, India and Madagascar, it has invaded urban and agriculture zones and has become a weed. The plant also causes breathing problems, dermatitis, and allergies, in susceptible people and animals (Towers et al., 1992; Dhawan et al., 1993). Insects and fungal pathogens from northeast Mexico have been assessed for their potential in the biological control of P. hysterophorus (Mcfayden, 1985; Evans 1987; Parker et al., 1994; McClay et al., 1995). In standard classical biological control, obligate parasites, in this case rust fungi (Uredinales) are the first choice, because they exhibit narrow host ranges, high reproductive capacities, and fast, efficient aerial dispersal (Evans and Ell ison, 1990). In the case of Parthenium the most promising fungal agents are: Puccinia abrupta var. parthenicola (Jackson) Parmelee (Uredinales), P. melampodii Diet. and Holw. (Uredinales), Plasmopara halstedii (Farlow) Berl. and De Toni (peronosporales), and Entyloma compositarum De Bary (Ustilaginales). Facultative pathogens of P. hysterophorus may also be appropiate fungal pathogens for biological control and could be mass-produced in the laboratory for future application in the field.

81 Seier and Romero (1996, 1997) carried out surveys in various regions of Mexico, in order to determine the fungal pathogens that could be useful as biocontrol agents of this weed, and collected nine mitosporic fungi associated with chlorotic and necrotic symptoms on old leaf tissues (Table 1). When a potential agent is sought for use in the biological control of a target weed, symptomatology helps in the selection process. Unspecialized pathogenic, mutualistic or endophytic fungi can however, produce symptoms similar to those of potential biological control agents. Stone et al. (1994) considered that endophytes have common attributes: I) they are internal, at least subticular, and have contact with and derive nutrition from the living host tissue; 2) they establish at least a transitory biotrophic nutritional relationship with their host; and 3) infected host tissues remain symptomless, i.e. disease free during their lifetime. Endophytes are also thought to be latent pathogens (Brown et al., 1998). Asymptomatic endophytes have diverse relationships with their host and may only produce disease symptoms under certain conditions. The purpose of this study is to identifY endophyte species that may exist as latent pathogens in P. hysterophorus, and evaluate methods to isolate these fungi from asymptomatic hosts.

Materials and methods Endophytes were isolated from leaves of mature healthy plants of Parthenium hysterophorus, from three sites in the central region ofthe state of Veracruz: Cotaxtla, Misantla and Emiliano Zapata. The vegetation ofCotaxtla was originally tropical rainforest, but is now mostly. mango plantations (Mangifera indica L.). Misantla was also a tropical forest, but has now been converted to grassland. Emiliano Zapata was a mixture of tropical deciduous forest and oak forest (Quercus spp.) and currently it is an agricultural area where vegetables are cultivated. P. hysterophorus is part of the native flora at all sites. Isolations were carried out in June and November 1996. Leaf surfaces were washed with distilled and sterilized water in order to remove epiphytic fungi. Seven methods of surface sterilisation (Schulz et al., 1993: Table 2) were utilised. Fifteen segments (5-10 mm2) were cut from leaves and used for each method, 105 segments were sterilized form each location and collection period. To stimulate the development of the mycelia, leaf segments were placed in Petri dishes with agar and malt extract (3.36%, Bioxon) and incubated at room temperature (~21 C). The mycelium developed within two weeks and the fungi were then transfered to new Petri dishes with the same media. The fungi were identified following sporulation.

82 Fungal Diversity

Results One-hundred and twenty-five endophyte strains were isolated from 420 leaf segments. Twenty-five "mycelia sterilia" (sensu Taylor et al., 1999) never sporulated, and eight had yeast-like growth (Table 3). Alternaria helianthi (Hansford) Tubaki and Nishihara was isolated from all sites. Alternaria alternata (Fr.) Keissler, Fusarium sp. and Penicilliumfuniculosum Thorn. were isolated from two sites. Nigrospora oryzae (Berk and Broome) Petch and Periconia sp. were iso lated from Site 1, Curvularia brachyspora Bredijn from Site 2, and, Cylindrocarpon sp. from Site 3 (Table 3). Fusarium sp. and Nigrospora oryzae ha\e note been recorded previously from this host (Table 1).

Table 1. Fungi previously recorded as pathogens from leaves of Parthenium hysterophorus in the state of Veracruz, Mexico (Seier and Romero, 1996, 1997). *Species also isolated as endophytic fungi.

Fungus *Alternaria alternata (Fr.) Keissler *A. helianthi (Hansford) Tubaki and Nishihara A. protenta Sirnrnons A. zinniae M.B. Ellis Cercospora sp. *Cylindrocarpon sp. *Penicillium funiculosum Thorn. *Periconia sp. Phoma subglomerata Boererna

Surface sterilization method I, yielded the largest number of endophytes (73 strains). However, it is possible that some epiphytes may also have been isolated with this method. Method III revealed the second largest number of endophytes (34 strains). Methods II and IV yielded 15 and three strains, respectively. In the case of methods V, VI and VII, where formaldhyde was used, no development ofmycelium was observed. Pathogens previously recorded from leaves of Parthenium hysterophorus are listed in the Table 1. Strains offive ofthe nine species live endophytically in tissues of P. hysterophorus (Table 2).

Discussion Most mitosporic fungi isolated as endophytes in this study also grow and sporulate on chlorotic and necrotic leaf tissues of P. hysterophorus. As some pathogens may have a latent phase within the host tissue, and some saprobes can also be facultative parasites, it may be that certain endophytes become

83 Table 2. Sterilization surface methods used with leaves of Parthenium hysterophorus (based on Schulz et al., 1993).

Method Solution Dilution with Time in distilled water solution I NaOCI (15%) 2:1 5 min II Ethanol 96% 2:1 1 min NaOCI 5 min Ethanol 96% 30 sec III Ethanol 96% 1 :3 30 sec Sterile water 30 sec NaOCI 5 min Ethanol 96% 30 sec Sterile water 2min IV Ethanol 96% 5 min V Formaldehyde 40% 1 min VI Formaldehyde 40% 3 min VII Formaldehyde 40% 5 min pathogenic when the host plant is stressed. Evidence suggests that endophytes have evolved directly from plant pathogenic fungi (Carrot, 1998; Isaac, 1992). A latent phase represents a specific condition where the fungus can either develop symptoms or cause changes in the physiology of the host plant. This condition may not be suitable in the biological control programme of weeds, because these fungi may rarely become pathogenic if the plant is healthy. Fungal biocontrol agents should have a rapid life cycles and cause damage to the host plants (Evans and Ellison, 1990). In other cases however, endophytic relationships represent an advantage in the control of pest and diseases in economically important plants or grasses (Clay, 1988; 1989). The host• endophyte association may result in protection against insects or other pathogens, instead of causing damage. Welty et al. (1991) observed that in the grass Festuca arundinacea Schreb the presence of endophytes reduce the incidence of Puccinia gram inis subsp. graminicola Z. Urban and Rhizoctonia zeae Voorhees (Gwinn and Gavin, 1992). Schardl and Phillips (1997) reported Epichloe and Neotyphodium species protecting various grasses against fungal pathogens, nematodes, insect and mammalian herbivores. It has been suggested that some endophytic species may be antagonistic towards other fungal species (Isaac, 1992). Because endophytes colonize the internal tissues of the host plant. This strategy assures the early occupation and possession of the resources, before the colonization by other fungi can occur, besides the production of alkaloids, terpenes and others, during symbiont stage. If this is the case the presence of endophytes in host plant may be problematical in

84 Fungal Diversity

biological control, because potential biological agents may be inhibited by the endophytes.

Table 3. Endophytic fungi isolated from leaves of Parthenium hysterophorus from three sites in the state of Veracruz, Mexico. Methods are the same as in Table 2; Ipreviously identified as pathogenic fungi (see Table I); ni = not isolated in this month; F = frequency.

mm1II Site Endophytes46Fm9183ni81214495F7nim1110(%)IntNovImIIIII1II(%) IVI1IIImniIIIIII Isolation method NigrosporaFusariumYeastUnidentifiedIlA.CurvularialA.IICylindrocarponIpericoniaIAlternariaPenicilliumalternataalternatahelianthisp. 21sp.sp.brachysporahelianthifuniculosumoryzaefuniculosumsp.sp. 246531sp. June Cotaxtla

Misantla

Emiliano Zapata

Our comparison of the surface sterilization methods, shows that methods I and III are most effective. With Method I we isolated most endophyte strains that produce spores. Method III, is probably used most often by reserchers (Petrini et al., 1992), and was more effective in isolating sterile mycelium and yeasts. Both methods can be used to obtain endophytic fungi in other herbaceus plants. Methods ITand IV were more or less specific to Fusarium sp. and A. alternata, but their frequency was low. Methods V, VI and VII seem to destroy the endophytes within the plant tissue. It is possible that these methods could be more useful if we reduced concentration and/or times of exposure to solutions. Schulz et al. (1993) obtained best results using method IV, in eleven ofthe twelve species of plants tested. Leaftissue characteristics vary with each plant species, thus several isolation methods should be tested in order to determine the optimum isolation protocol. There is however, a risk in isolating

85 pathogenic strains, either when the pathogen is in an incubation phase or it is latent within the host tissue. It is also possible that some of the fungi isolated as endophytes, are epiphytes. An understanding of endophytes present in the target weed should prove to be valuable in biological control regimes. This may increase the possibility of success of the agents selected. If endophytic relationships of potential biocontrol agents are recognized within host plants, or if the establishment of biocontrol agents are affected by endophytes inside leaf tissues, some pathogenic fungi may have to be discarded as potential biocontrol agents.

Acknowledgements We gratefully acknowledge financial support from the MRC. The IIBC/CSIRO programme of Biological Control of Parthenium weed, and Instituto de Ecologia, AC. Nos. 902-XX and 902-16. We would like to thank J.F. White, The State University of New Jersey for critically reviewing the manuscript.

References

Brown, KB., Hyde, KD. and Guest, DJ. (1998). Preliminary studies on endophytic fungal communities of Musa acuminata species complex in Hong Kong and Australia. Fungal Diversity 1: 27-51. Carrol, G. (1988). Fungal endophytes in stems and leaves: from latent pathogen to mutualistic symbiont. Ecology 69: 2-9. Clay, K (1988). Fungal endophytes of grasses: a defensive mutualism between plants and fungi. Ecology 69: 10-16. Clay, K (1989). Clavicipitaceaous endophytes of grasses: their potential as biocontrol agents. Mycological Research 92: 1-12. Dhawan, S.R., Aneja, K:R. and Dhawan, P. (1993). Parthenium hysterophorus L., the danger weed and its control. Biome 6: 117-122. Evans, RC. (1987). Life-cycle of Puccinia abrupta var. partheniicola, a potential biological control agent of Parthenium hysterophorus. Transactions of the British Mycological Society 88: 105-111. Evans, H.C and Ellison, C.A (1990). Classical biological control of weeds with micro• organisms: past, present, prospects. Aspects of Applied Biology 24: 39-49. Gwinn, K.D. and Gavin, AM. (1992). Relationship between endophyte infestation level of tall fescue seed lots and Rhizoctonia zeae seedling disease. Plant Disease 76: 911-914. Isaac, S. (1992). Fungal-Plant Interactions. Chapman and Hall, New York. McClay, AS., Palmer, W.A., Bennett, F.D. and Pullen, KR. (1995). Phytophagous arthropods associated with Parthenium hysterophorus (Asteraceae) in North America. Environmental Entomology 24: 796-809. McFadyen, R.E. (1985). The biological control programme against Parthenium hysterophorus in Queensland. In: Proceedings VI International Symposium on the Biol.ogical Control of Weeds, August, 19-25, 1984, Vancouver, Canada (E.S. Delfosse, ed). Agriculture Canada pp 789-796. Nash, D.L. and Williams, L.O. (1976). Flora of Guatemala. Fieldiana Botanica 24: 664-665.

86 Fungal Diversity

Parker, A., Holden, A.N.G. and Tomley, AJ. (1994). Host specificity testing and assessment of the pathogenicity of the rust, Puccinia abrupta var. partheniicola, as a biological control agent of parthenium weed (Parthenium hysterophorus). Plant Pathology 43: 1• 16. Petrini, 0., Sieber, T.N., Toti, L. and Viret, O. (1992). Ecology, metabolite production, and substrate utilization in endophytic fungi. Natural Toxins 1: 185-196. Schardl, C.L. and Phillips, T.D. (1997). Protective grass endophytes. Where are they from and where are they going. Plant Disease 81: 430-438. Schulz, B., Wanke, U., Draeger, S. and Aust, HJ. (1993). Endophytes from herbaceous plants and shrubs: effectivenses of surfuse sterilization methods. Mycological Research 97: 1447-1450. Seier, M. and Romero, A. (1996). Control of parthenium weed (Parthenium hysterophorus), July, 1995-June 1996. Annual Report, IlBC. Seier, M. and Romero, A. (1997). Control of parthenium weed (Parthenium hysterophorus), July, 1996-June 1997. Annual report, IIBe. Stone, lK, Viret, 0., Petrini O. and Chapela, I.H. (1994). Histological studies of host penetration and colonization by endophytic fungi. In: Host wall alterations by parasitic fungi (eds O. Petrini and G.B. Oullette). APS Press, USA: 115-126. Taylor, J.E., Hyde, K.D. and Jones, ERG. (1999). Endophytic fungi associated with the temperate palm Trachycarpus fortunei within and outside its natural geographic range. New Phytologist 142: 335-346. Towers, G.H.N., Mitchell, T.C. Rodriguez, E. Bennett, F.D. and Subba Roa, P.V. (1992). Biology and chemistry of Parthenium hysterophorus L., a problem weed in India. Journal Science of India Research 36: 672-684. Welty, R.E., Barker, R.E. and Azevedo, M.D. (1991). Reaction of tall infected and noninfected by Acremoniun coenophialum to Puccinia graminis subsp. graminicola. Plant Disease 75: 883-886.

(Received 27 March 2000, accepted 16 April 2001)

87